Tencent Hunyuan released Hy4, the successor to Hy3, scoring 35 on the Artificial Analysis Intelligence Index v5 — tied with Grok 4.7 and one point behind Claude Haiku 5.5, ranking #2 among China-based labs behind Kimi K3.1 Pro's 46. Hy4 leads Chinese models on agentic coding (31% on FrontierCode 2.0) but shows an 89% hallucination rate on AA-Omniscience versus Kimi's 37%.
Cache Read
缓存读取
Retrieving data from pre-stored context caches to reduce inference latency and cost.
Curated Intelligence
Anthropic released Claude Haiku 5.5, reducing average operational costs by ~75% compared to its predecessor, with input/output prices starting at $0.10/$0.50 per million tokens. It is the first Haiku model to support adjustable performance settings and is explicitly positioned as a sub-agent for coding tasks alongside Opus and Sonnet.
Anthropic released Claude Haiku 5.5, cutting average costs by 75% and input prices by up to 90% for short prompts. The company also slashed pricing for Sonnet 5.5 and updated SDKs to support computer and browser use.
Anthropic released Claude Haiku 5.5, a small model optimized for high-volume tasks with costs reduced by ~75% compared to Haiku 4.5, serving as a coding subagent for larger models. Additionally, Sonnet 5.5's cache read prices were halved, lowering the cost of most agentic workloads by ~20%.
OpenAI has released GPT-6 Sol and Luna, cutting API prices by approximately 50% while maintaining intelligence levels comparable to the previous GPT-5.6 generation. By optimizing inference efficiency, these models significantly reduce per-task costs (e.g., Sol drops to $1.06/task), reinforcing OpenAI's position on the cost-efficiency Pareto frontier. Although coding agent performance shows divergence—with Sol improving but Luna regressing in certain benchmarks—the substantial leap in value-for-money makes them critical for developers deploying high-load applications.